Guangdong Deploys AI and Satellite Mapping to Streamline Crop Insurance

Guangdong's provincial surveying institute has integrated satellite remote sensing, high-precision positioning, and artificial intelligence to modernize agricultural insurance workflows. According to a report by People's Daily, the system automates crop verification during policy underwriting and delivers disaster damage assessments within hours of severe weather events such as typhoons. Supported by the 605-station GDCORS positioning network and commercial satellite partnerships, the initiative reflects a broader effort by Chinese natural resources authorities to deploy geospatial data infrastructure for rural risk management and economic governance.

Replacing manual, parcel-by-parcel field inspections with automated satellite AI processing reduces damage assessment timelines from weeks to hours following typhoons and floods. This accelerates payouts, minimizes subjective estimation disputes between farmers and insurers, and demonstrates the practical utility of public spatial infrastructure—such as continuous operational reference systems—in commercial and policy insurance markets.

Agricultural insurance in southern China has long grappled with slow, labor-intensive claims verification following seasonal typhoons. When storms batter coastal farmland, claims adjusters previously had to survey thousands of acres on foot—a process that could take days or weeks, sometimes running past harvest time. To address this bottleneck, the Guangdong Institute of Land and Resources Surveying and Mapping has developed a spatial information workflow that automates assessments, according to a report by People's Daily. Following a typhoon, high-resolution satellite imagery is transmitted within hours. Machine-learning algorithms automatically compare pre- and post-disaster imagery to delineate damaged parcels and calculate the severity of loss, pushing standardized loss reports directly to insurers' back-end systems. Xiao Jianneng, an engineer at the institute, told People's Daily that the service has operated since 2018 and spans three stages of the insurance lifecycle: underwriting, growth monitoring, and damage evaluation. During underwriting, satellite feeds and farmland boundary datasets establish a precise baseline, automatically classifying crop varieties to prevent policy misallocations. In policy-backed rice insurance, for instance, exact boundary extraction prevents under- or over-insuring plots. During the crop cycle, multi-temporal remote-sensing data tracks vegetation health. Because compensation standards depend heavily on crop maturity—losses are far lower during the seedling stage than immediately prior to harvest—the imagery provides actuarial benchmarks for tailored payouts, Xiao explained. When natural disasters hit, rapid satellite access combined with low-altitude drone surveys enables fast damage quantification. Xiao noted that the institute relies on domestic public satellites, inter-provincial data sharing, and long-term contracts with commercial satellite providers to guarantee fast tasking. The agricultural platform draws on wider provincial geospatial infrastructure. Zhou Xubin, head of the institute, stated that Guangdong operates the GDCORS continuous operational reference system, which comprises 605 reference stations. The network delivers real-time centimeter-level positioning to more than 1,800 surveying entities and nearly 100,000 practitioners, processing approximately 400 million service responses annually. Authorities are now leveraging this underlying network alongside aviation, drones, and ground terminals to expand spatial intelligence across other regional governance sectors.